Skip to content

Pull requests: google/XNNPACK

New pull request New
145 Open 7,756 Closed
145 Open 7,756 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Enable f32 packw scalar microkernel when using scalar or wasm GEMM
#8177 opened Apr 1, 2025 by copybara-service bot Loading…
- Using the output to get the max value is risky. If the output produces junk data, the max value could be infinity, and then the test will pass with any output.
#8176 opened Apr 1, 2025 by copybara-service bot Loading…
GEMM HVX use qfloat accumulators
#8168 opened Mar 31, 2025 by copybara-service bot Loading…
Add support for blockwise quantization with fp16 scales.
#8161 opened Mar 31, 2025 by copybara-service bot Loading…
Reformat C too
#8155 opened Mar 31, 2025 by copybara-service bot Loading…
Internal change only.
#8151 opened Mar 28, 2025 by copybara-service bot Loading…
Copybara import of the project:
#8143 opened Mar 27, 2025 by copybara-service bot Loading…
Add scalar packing kernels for bf16_f32 batch matmuls
#8136 opened Mar 27, 2025 by copybara-service bot Loading…
Disable support for padding in unpooling.
#8117 opened Mar 24, 2025 by copybara-service bot Loading…
[RVV] Add qu8-gemm/qu8-igemm kernels for rvv
#8105 opened Mar 23, 2025 by ken-unger Loading…
3
Support Sparse Inference for RVV
#8081 opened Mar 20, 2025 by NonerKao Loading…
4
Add XNN_FLAG_CONSISTENT_ARITHMETIC
#8077 opened Mar 19, 2025 by copybara-service bot Loading…
Enable sin and cos delegation for XNNPACK.
#8057 opened Mar 17, 2025 by copybara-service bot Loading…
Make the microkernel header files more self-contained, i.e. include the headers they require.
#8037 opened Mar 14, 2025 by copybara-service bot Loading…
Extend bf16 gemm config.
#8027 opened Mar 13, 2025 by copybara-service bot Loading…
Minor updates to xnnpack.h comments
#8018 opened Mar 12, 2025 by copybara-service bot Loading…
Move operator API to internal header
#8017 opened Mar 12, 2025 by copybara-service bot Loading…
Implemented F32-GEMM via SIMD wrappers.
#8003 opened Mar 11, 2025 by copybara-service bot Loading…
1
Fix compiler error when trying to use f16-neonfp16arith
#7889 opened Feb 26, 2025 by copybara-service bot Loading…
Enable support for Emscripten in CMake
#7859 opened Feb 21, 2025 by aiusepsi Loading…
3
CMake enable AMX on clang/gcc 11
#7850 opened Feb 19, 2025 by copybara-service bot Loading…
Move assembly comment about GP registers being saved
#7837 opened Feb 18, 2025 by copybara-service bot Loading…
Adding support for SME1 GEMM FP32 kernel
#7831 opened Feb 18, 2025 by vgundlur Loading…
3
Enable AVXVNNI microkernels for mobile platforms
#7828 opened Feb 18, 2025 by soniravi31 Loading…
3
Add QNX to recognized systems
#7798 opened Feb 10, 2025 by eleir9268 Loading…
5
ProTip! Filter pull requests by the default branch with base:master.